10
Figure 15. Phase Response Figure 16. Propagation Delay vs Temperature
Figure 18. Input to Output Propagation Delay Timing Diagram. V
Out Di =
V
Out+ -
V
Out-
Figure 17. Shutdown And Wakeup Input To Output Timing Diagram. V
Out Di =
V
Out+ -
V
Out-
Vin
V
SD
V
Ou t D i 
+2 V
-2.46 V
0 V
0 V
2 V
0 V
5V
t
SD
t
ON
V
IN
V
Out Di
0 V
2 V
0 V
2 V
T
PLH50-10
T
PLH50-50
T
PLH50-90
TPLH 50-10
TPLH 50-50
TPLH 50-90
0
20
40
60
80
100
120
140
160
180
200
1000 10000 100000
Phase (deg)
Bandwidth (Hz)
0
1
2
3
4
5
6
-55 -35 -15 5 25 45 65 85 105 125
Prog Delay (PS)
Temp (qC)
11
Figure 19. Typical application circuit.
De nitions
Gain
Gain is de ned as the slope of the best- t line of di eren-
tial output voltage (V
OUT+
V
OUT-
) over the nominal input
range, with o set error adjusted out.
Nonlinearity
Nonlinearity is de ned as half of the peak-to-peak output
deviation from the best- t gain line, expressed as a per-
centage of the full-scale di erential output voltage.
Common Mode Transient Immunity, CMTI, also known
as Common Mode Rejection
CMTI is tested by applying an exponentially rising/falling
voltage step on pin 4 (GND1) with respect to pin 5 (GND2).
The rise time of the test waveform is set to approximately
50 ns. The amplitude of the step is adjusted until the dif-
ferential output (V
OUT+
V
OUT-
) exhibits more than a 200
mV deviation from the average output voltage for more
than 1µs. The ACPL-C87x will continue to function if more
than 10 kV/s common mode slopes are applied, as long
as the breakdown voltage limitations are observed.
Power Supply Rejection, PSR
PSRR is the ratio of di erential amplitude of the ripple
outputs over power supply ripple voltage, referred to the
input, expressed in dB.
Application Information
Application Circuit
The typical application circuit is shown in Figure 19.
The ACPL-C87X voltage sensor is often used in photo-
voltaic (PV) panel voltage measurement and tracking in
PV inverters, and DC bus voltage monitoring in motor
drivers. The high voltage across rails needs to be scaled
down to  t the input range of the iso-amp by choosing R1
and R2 values according to appropriate ratio.
The ACPL-C87X senses the single-ended input signal
and produces di erential outputs across the galvanic
isolation barrier. The di erential outputs (Vout+, Vout-)
can be connected to an op-amp to convert to a single-
ended signal or directly to two ADCs. The op-amp used in
the external post-ampli er circuit should be of su ciently
high precision so that it does not contribute a signi cant
amount of o set or o set drift relative to the contribu-
tion from the isolation ampli er. Generally, op-amps with
bipolar input stages exhibit better o set performance
than op-amps with JFET or MOSFET input stages.
In addition, the op-amp should also have enough
bandwidth and slew rate so that it does not adversely
a ect the response speed of the overall circuit. The post-
ampli er circuit includes a pair of capacitors (C4 and C5)
that form a single-pole low-pass  lter; these capacitors
allow the bandwidth of the post-amp to be adjusted in-
dependently of the gain and are useful for reducing the
output noise from the isolation ampli er.
The gain-setting resistors in the post-amp should have a
tolerance of 1% or better to ensure adequate CMRR and
adequate gain tolerance for the overall circuit. Resistor
networks can be used that have much better ratio toler-
ances than can be achieved using discrete resistors. A
resistor network also reduces the total number of compo-
nents for the circuit as well as the required board space.
V
DD1
1
V
IN
2
SHDN3
GND14 GND2 5
V
OUT-
6
V
OUT+
7
V
DD2
8
U1
ACPL-C87X
GND2
R4
10K,1%
GND2
V
DD2
V
DD1
Vout
GND1
L1
L2
R2
10K
C1
100 pF
C2
100 nF
C3
100 nF
R3
10K,1%
R1
U2
OPA237
V+
V-
C4
100 pF
R5
10K, 1%
C5
100 pF
R6
10K, 1%
12
The input stage of the typical application circuit in Figure
19 can be simpli ed as the diagram shown in Figure 20.
R2 and R
IN
, input resistance of the ACPL-C87X, create a
current divider that results in an additional measurement
error component that will add on to the tot on top of the
device gain error. With the assumption that R1 and R
IN
have a much higher value than R2, the resulting error can
be estimated to be R2/R
IN
.
With R
IN
of 1 G for the ACPL-C87X, this additional mea-
surement error is negligible with R2 up to 1 M, where the
error is approximately 0.1%. Though small, it can be further
reduced by reducing the R2 to 100 k (error of 0.01%
approximately), or 10 k (error of 0.001% approximately).
However with lower R2, a drawback of higher power dis-
sipation in the resistive divider string needs to be consid-
ered, especially in higher voltage sensing applications. For
example, with 600 V DC across L1 and L2 and R2 of 100 k
for 0.01% measurement error, the resistive divider string
Figure 20. Simpli ed Input Stage.
Figure 21. Thermistor sensing in IGBT Module
consumes about 12 mW, assuming V
IN
is set at 2 V. If the R2
is reduced to 10 k to reduce error to 0.001%, the power
consumption will increase to about 120 mW. In energy
e ciency critical applications such as PV inverters and
battery-powered applications, this trade-o between
measurement accuracy and power dissipation in the
resistive string provides  exibility in design priority.
Isolated Temperature Sensing using Thermistor
IGBTs are an integral part of a motor or servo drive system
and because of the high power that they usually handle,
it is essential that they have proper thermal management
and are su ciently cooled. Long term overload conditions
could raise the IGBT module temperature permanently or
failure of the thermal management system could subject
the module to package overstress and lead to catastrophic
failures. One common way to monitor the temperature
of the module is through using a NTC type thermistor
mounted onto the IGBT module. Some IGBT module man-
ufacturers also have IGBTs that comes with the thermistor
integrated inside the module. In some cases, it is necessary
to isolate this thermistor to provide added isolation and
insulation due to the high power nature of the IGBTs. The
ACPL-C87x voltage sensor can be used to easily meet
such a requirement, while providing good accuracy and
non-linearity. Figure. 21 shows an example of such an
implementation. The ACPL-C87x is used to isolate the
thermistor voltage which is later fed by the post amp
stage to an ADC onboard the microcontroller (MCU) to
determine the module temperature. The thermistor needs
to be biased in way that its voltage output will optimize
the 2 V input range of the ACPL-C87x across the intended
temperature measurement range.
Measurement Accuracy and Power Dissipation of the Resistive Divider
+ –
+
GND
ACPL-C87x
R1
R2
R
IN
HV+
HV-
IGBT Module
NTC Thermistor
U
V
W
+
+
GND
ACPL-C87x
Vdd
MCU
ADC
Post
Amp

ACPL-C87B-000E

Mfr. #:
Manufacturer:
Broadcom / Avago
Description:
Optically Isolated Amplifiers Precision Iso-Amp
Lifecycle:
New from this manufacturer.
Delivery:
DHL FedEx Ups TNT EMS
Payment:
T/T Paypal Visa MoneyGram Western Union